Abstract
In this article, we survey existing academic and commercial efforts to provide Field-Programmable Gate Array (FPGA) acceleration in datacenters and the cloud. The goal is a critical review of existing systems and a discussion of their evolution from single workstations with PCI-attached FPGAs in the early days of reconfigurable computing to the integration of FPGA farms in large-scale computing infrastructures. From the lessons learned, we discuss the future of FPGAs in datacenters and the cloud and assess the challenges likely to be encountered along the way. The article explores current architectures and discusses scalability and abstractions supported by operating systems, middleware, and virtualization. Hardware and software security becomes critical when infrastructure is shared among tenants with disparate backgrounds. We review the vulnerabilities of current systems and possible attack scenarios and discuss mitigation strategies, some of which impact FPGA architecture and technology. The viability of these architectures for popular applications is reviewed, with a particular focus on deep learning and scientific computing. This work draws from workshop discussions, panel sessions including the participation of experts in the reconfigurable computing field, and private discussions among these experts. These interactions have harmonized the terminology, taxonomy, and the important topics covered in this manuscript.
- [1] . 2018. DLA: Compiler and FPGA overlay for neural network inference acceleration. In 28th International Conference on Field Programmable Logic and Applications (FPL). IEEE, 411–4117.Google ScholarCross Ref
- [2] . 2017. Implementation of virtual network function chaining through segment routing in a Linux-based NFV infrastructure. In IEEE Conference on Network Softwarization (NetSoft). 1–5.
DOI: https://doi.org/10.1109/NETSOFT.2017.8004208Google ScholarCross Ref - [3] . 2017. An FPGA platform for hyperscalers. In IEEE 25th Annual Symposium on High-Performance Interconnects. 29–32.
DOI: https://doi.org/10.1109/HOTI.2017.13Google Scholar - [4] . 2019. Cloud-based FPGA custom computing machines for streaming applications. IEEE Access 7 (2019), 38009–38019.Google ScholarCross Ref
- [5] . 2019. RAM-Jam: Remote temperature and voltage fault attack on FPGAs using memory collisions. In Workshop on Fault Diagnosis and Tolerance in Cryptography. 48–55.Google Scholar
- [6] . 2020. Tackling hardware/software co-design from a database perspective. In 6th Biennial Conference on Innovative Data Systems Research (CIDR).Google Scholar
- [7] 2021. Amazon EC2 F1 Instances. Retrieved from https://aws.amazon.com/ec2/instance-types/f1/.Google Scholar
- [8] . 1995. Teramac-configurable custom computing. In IEEE Symposium on FPGAs for Custom Computing Machines. 32–38. Google ScholarDigital Library
- [9] . 2010. AMBA 4 AXI4-Stream Protocol Specification.
Technical Report . ARM.Google Scholar - [10] . 2011. AMBA AXI and ACE Protocol Specification.
Technical Report . ARM.Google Scholar - [11] . 2017. Virtualized execution runtime for FPGA accelerators in the cloud. IEEE Access 5 (2017), 1900–1910.Google ScholarCross Ref
- [12] . 1999. MPLS and traffic engineering in IP networks. IEEE Commun. Mag. 37, 12 (1999), 42–47. Google ScholarDigital Library
- [13] . 2011. Towards predictable datacenter networks. In ACM SIGCOMM Conference (SIGCOMM’11). Association for Computing Machinery, New York, NY, 242–253.
DOI: https://doi.org/10.1145/2018436.2018465Google ScholarDigital Library - [14] . 2007. Maxwell – A 64 FPGA supercomputer. In 2nd NASA/ESA Conference on Adaptive Hardware and Systems (AHS’07). 287–294.
DOI: https://doi.org/10.1109/AHS.2007.71 Google ScholarDigital Library - [15] . 2010. Short-circuits on FPGAs caused by partial runtime reconfiguration. In International Conference on Field Programmable Logic and Applications. 596–601. Google ScholarDigital Library
- [16] . 2014. Containers and cloud: From LXC to Docker to Kubernetes. IEEE Cloud Comput. 1, 3 (2014), 81–84.Google ScholarCross Ref
- [17] . 2021. Toward multi-FPGA acceleration of the neural networks. ACM J. Emerg. Technol. Comput. Syst. 17, 2 (2021), 1–23. Google ScholarDigital Library
- [18] . 2013. Achieving 10Gbps line-rate key-value stores with FPGAs. In 5th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud’13). USENIX Association, San Jose, CA. Retrieved from https://www.usenix.org/conference/hotcloud13/workshop-program/presentations/blott.Google Scholar
- [19] . 2015. Scaling out to a single-node 80gbps memcached server with 40terabytes of memory. In 7th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage’15). USENIX Association, Santa Clara, CA. Retrieved from https://www.usenix.org/conference/hotstorage15/workshop-program/presentation/blott. Google ScholarDigital Library
- [20] . 2018. FINN-R: An end-to-end deep-learning framework for fast exploration of quantized neural networks. ACM Trans. Reconfig. Technol. Syst. 11, 3 (2018), 1–23. Google ScholarDigital Library
- [21] . 2007. Introduction to Reconfigurable Computing: Architectures, Algorithms, and Applications (1st ed.). Springer Publishing Company, Incorporated. Google ScholarDigital Library
- [22] . 2005. The Erlangen slot machine: Increasing flexibility in FPGA-based reconfigurable platforms. In IEEE International Conference on Field-Programmable Technology.37–42.
DOI: https://doi.org/10.1109/FPT.2005.1568522 Google ScholarDigital Library - [23] . 2008. Automata-based assertion-checker synthesis of PSL properties. ACM Trans. Des. Autom. Electron. Syst. 13, 1 (
Feb. 2008).DOI: https://doi.org/10.1145/1297666.1297670 Google ScholarDigital Library - [24] . 2020. Beyond peak performance: Comparing the real performance of AI-Optimized FPGAs and GPUs. In International Conference on Field-Programmable Technology (ICFPT). 10–19.
DOI: https://doi.org/10.1109/ICFPT51103.2020.00011Google ScholarCross Ref - [25] . 2012. ZUMA: An open FPGA overlay architecture. In IEEE 20th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM). IEEE, 93–96. Google ScholarDigital Library
- [26] . 2014. FPGAs in the cloud: Booting virtualized hardware accelerators with openstack. In IEEE 22nd Annual International Symposium on Field-Programmable Custom Computing Machines. IEEE, 109–116. Google ScholarDigital Library
- [27] . 2016. A cloud-scale acceleration architecture. In 49th IEEE/ACM International Symposium on Microarchitecture. 1–13. Google ScholarDigital Library
- [28] . 2013. An FPGA memcached appliance. In ACM/SIGDA International Symposium on Field Programmable Gate Arrays. 245–254. Google ScholarDigital Library
- [29] . 2014. Enabling FPGAs in the cloud. In 11th ACM Conference on Computing Frontiers. ACM, 3. Google ScholarDigital Library
- [30] . 2021. ThunderGP: HLS-based graph processing framework on FPGAs. In ACM/SIGDA International Symposium on Field-Programmable Gate Arrays. 69–80. Google ScholarDigital Library
- [31] . 2016. When Spark meets FPGAs: A case study for next-generation DNA sequencing acceleration. In 8th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud’16). Google ScholarDigital Library
- [32] . 2014. Map-Reduce processing of k-means algorithm with FPGA-accelerated computer cluster. In IEEE 25th International Conference on Application-Specific Systems, Architectures and Processors. 9–16.Google ScholarCross Ref
- [33] . 2018. Serving DNNs in real time at datacenter scale with project brainwave. IEEE Micro 38, 2 (2018), 8–20.Google ScholarCross Ref
- [34] . 2020. Vast majority of cyber-attacks on cloud servers aim to mine cryptocurrency. Retrieved from https://www.zdnet.com/article/vast-majority-of-cyber-attacks-on-cloud-servers-aim-to-mine-cryptocurrency/.Google Scholar
- [35] . 2005. NARC: Network attached reconfigurable computing for high performance, network based applications. In 8th Annual International Conference on Military and Aerospace Programmable Logic Devices (MAPLD’05).Google Scholar
- [36] . 2014. Online scheduling for FPGA computation in the cloud. In International Conference on Field-Programmable Technology (FPT). IEEE, 330–333.Google ScholarCross Ref
- [37] . 2019. Streaming message interface: High-performance distributed memory programming on reconfigurable hardware. In International Conference for High Performance Computing, Networking, Storage and Analysis. 1–33. Google ScholarDigital Library
- [38] . 2019. FPGA-accelerated machine learning inference as a service for particle physics computing. Comput. Softw. Big Sci. 3, 1 (2019), 1–15.Google ScholarCross Ref
- [39] . 2021. Leaky buddies: Cross-component covert channels on integrated CPU-GPU systems. In ACM/IEEE 48th Annual International Symposium on Computer Architecture (ISCA). IEEE, 972–984. Google ScholarDigital Library
- [40] . 2012. FPGAs for trusted cloud computing. In 22nd International Conference on Field Programmable Logic and Applications (FPL). 63–70.Google ScholarCross Ref
- [41] . 2020. The unpatchable silicon: A full break of the bitstream encryption of Xilinx 7-series FPGAs. In 29th USENIX Security Symposium.Google Scholar
- [42] . 2019. NICA: An infrastructure for inline acceleration of network applications. In USENIX Annual Technical Conference (USENIX ATC 19). 345–362. Google ScholarDigital Library
- [43] . 2020. Competing in Artificial Intelligence Chips: China’s Challenge Amid Technology War. Centre for International Governance Innovation, Special Report.Google Scholar
- [44] . 2019. A modular heterogeneous stack for deploying FPGAs and CPUs in the data center. In ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA’19). ACM, New York, NY, 262–271.
DOI: https://doi.org/10.1145/3289602.3293909 Google ScholarDigital Library - [45] . 2015. Virtualized FPGA accelerators for efficient cloud computing. In IEEE 7th International Conference on Cloud Computing Technology and Science (CloudCom). IEEE, 430–435. Google ScholarDigital Library
- [46] . 2017. Secure function evaluation using an FPGA overlay architecture. In ACM/SIGDA International Symposium on Field-Programmable Gate Arrays. 257–266. Google ScholarDigital Library
- [47] . 2018. Azure accelerated networking: SmartNICs in the public cloud. In 15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18). 51–66. Google ScholarDigital Library
- [48] . 2018. Xilinx FPGAs: The Chip Behind Alibaba’s Singles Day. Retrieved from https://www.forbes.com/sites/moorinsights/2018/11/29/xilinx-fpgas-the-chip-behind-alibabas-singles-day/?sh=5f2294e27e3b.Google Scholar
- [49] . 2018. A configurable cloud-scale DNN processor for real-time AI. In ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA). 1–14.
DOI: https://doi.org/10.1109/ISCA.2018.00012 Google ScholarDigital Library - [50] . 2017. Google Cloud TPU: Strategic Implications for Google, NVIDIA and the Machine Learning Industry. Retrieved from https://www.forbes.com/sites/moorinsights/2017/05/22/google-cloud-tpu-strategic-implications-for-google-nvidia-and-the-machine-learning-industry/?sh=69d2f5a13af7.Google Scholar
- [51] . 2019. Parallel processing on FPGA combining computation and communication in OpenCL programming. In IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW). IEEE, 479–488.Google Scholar
- [52] . 2019. Dynamic application reconfiguration on heterogeneous hardware. In 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments. 165–178. Google ScholarDigital Library
- [53] . 2020. How Arm Came to Dominate the Mobile Market and It’s Coming for More, Much More. Retrieved from https://www.techspot.com/article/1989-arm-inside.Google Scholar
- [54] . 2010. Novo-G: At the forefront of scalable reconfigurable supercomputing. Comput. Sci. Eng. 13, 1 (2010), 82–86. Google ScholarDigital Library
- [55] . 2016. Novo-G#: Large-scale reconfigurable computing with direct and programmable interconnects. In IEEE High Performance Extreme Computing Conference (HPEC). 1–7.
DOI: https://doi.org/10.1109/HPEC.2016.7761639Google ScholarCross Ref - [56] . 2016. Accelerating Apache Spark big data analysis with FPGAs. In International IEEE Conferences on Ubiquitous Intelligence Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People, and Smart World Congress (UIC/ATC/ScalCom/CBDCom/IoP/SmartWorld). 737–744.
DOI: https://doi.org/10.1109/UIC-ATC-ScalCom-CBDCom-IoP-SmartWorld.2016.0119Google Scholar - [57] . 2019. Accelerating Apache Spark with FPGAs. Concurr. Comput.: Pract. Exper. 31, 2 (2019), e4222.
DOI: https://doi.org/10.1002/cpe.4222arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1002/cpe.4222 .Google ScholarCross Ref - [58] . 2019. Reading between the dies: Cross-SLR covert channels on multi-tenant cloud FPGAs. In IEEE International Conference on Computer Design. 1–10.Google ScholarCross Ref
- [59] . 2020. CAPSULe: Cross-FPGA covert-channel attacks through power supply unit leakage. In IEEE Symposium on Security and Privacy. 1728–1741.Google Scholar
- [60] . 2018. Leaky wires: Information leakage and covert communication between FPGA long wires. In Asia Conference on Computer and Communications Security (ASIACCS). 15–27. Google ScholarDigital Library
- [61] . 2020. AFI Power. Retrieved from https://github.com/aws/aws-fpga/blob/master/hdk/docs/afi_power.md.Google Scholar
- [62] . 2017. Voltage drop-based fault attacks on FPGAs using valid bitstreams. In International Conference on Field Programmable Logic and Applications (FPL). 1–7.Google ScholarCross Ref
- [63] . 2021. Heterogeneous computing systems for complex scientific discovery workflows. In Design, Automation Test in Europe Conference Exhibition (DATE). 13–18.
DOI: https://doi.org/10.23919/DATE51398.2021.9474061Google ScholarCross Ref - [64] . 2021. OCT: The open cloud FPGA testbed. In 31st International Conference on Field Programmable Logic and Applications (FPL).Google Scholar
- [65] . 2001. Coarse grain reconfigurable architecture (embedded tutorial). In Asia and South Pacific Design Automation Conference. ACM, 564–570. Google ScholarDigital Library
- [66] . 2018. Inheriting software security policies within hardware IP components. In IEEE International Symposium on Field-Programmable Custom Computing Machines. 53–56.
DOI: https://doi.org/10.1109/FCCM.2018.00017Google Scholar - [67] . 2018. Secure hardware kernels execution in CPU+ FPGA heterogeneous cloud. In International Conference on Field-Programmable Technology (FPT). IEEE, 182–189.Google ScholarCross Ref
- [68] . 2017. First in-depth look at Google’s new second-generation TPU. Retrieved from https://www.nextplatform.com/2017/05/17/first-depth-look-googles-new-second-generation-tpu/.Google Scholar
- [69] . 2002. Dynamic hardware plugins in an FPGA with partial run-time reconfiguration. In 39th Annual Design Automation Conference (DAC’02). Association for Computing Machinery, New York, NY, 343–348.
DOI: https://doi.org/10.1145/513918.514007 Google ScholarDigital Library - [70] . 2019. Garbled circuits in the cloud using FPGA enabled nodes. In IEEE High Performance Extreme Computing Conference (HPEC). 1–6.Google ScholarCross Ref
- [71] . 2008. Designing secure systems on reconfigurable hardware. ACM Trans. Des. Autom. Electron. Syst. 13, 3 (
July 2008), 44:1–44:24. Google ScholarDigital Library - [72] . 2018. Maxelerator: FPGA accelerator for privacy preserving multiply-accumulate (MAC) on cloud servers. In 55th Annual Design Automation Conference. 1–6. Google ScholarDigital Library
- [73] . 2016. Consensus in a box: Inexpensive coordination in hardware. In 13th USENIX Symposium on Networked Systems Design and Implementation (NSDI’16). 425–438. Google ScholarDigital Library
- [74] . 2020. Security of cloud FPGAs: A survey. arxiv arXiv:2005.04867 (2020).Google Scholar
- [75] . 2017. Terabyte sort on FPGA-accelerated flash storage. In IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM). 17–24.
DOI: https://doi.org/10.1109/FCCM.2017.53Google Scholar - [76] . 2014. FlowBender: Flow-level adaptive routing for improved latency and throughput in datacenter networks. In 10th ACM International Conference on Emerging Networking Experiments and Technologies (CoNEXT’14). Association for Computing Machinery, New York, NY, 149–160.
DOI: https://doi.org/10.1145/2674005.2674985 Google ScholarDigital Library - [77] . 2015. HopLite: Building austere overlay NOCs for FPGAs. In 25th International Conference on Field Programmable Logic and Applications (FPL). IEEE, 1–8.Google ScholarCross Ref
- [78] . 2018. Sharing, protection, and compatibility for reconfigurable fabric with Amorphos. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI’18). 107–127. Google ScholarDigital Library
- [79] . 2014. Flipping bits in memory without accessing them: An experimental study of DRAM disturbance errors. In International Symposium on Computer Architecture. 361–372. Google ScholarDigital Library
- [80] . 2018. OpenCL-ready high speed FPGA network for reconfigurable high performance computing. In International Conference on High Performance Computing in Asia-Pacific Region. 192–201. Google ScholarDigital Library
- [81] . 2019. Spectre attacks: Exploiting speculative execution. In 40th IEEE Symposium on Security and Privacy (S&P’19).Google Scholar
- [82] . 2020. Do OS abstractions make sense on FPGAs? In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI’20). 991–1010. Google ScholarDigital Library
- [83] . 2020. FPGA accelerated embedded system security through hardware isolation. In Asian Hardware Oriented Security and Trust Symposium (AsianHOST). 1–6.
DOI: https://doi.org/10.1109/AsianHOST51057.2020.9358258Google Scholar - [84] . 2021. Compiler-driven FPGA virtualization with SYNERGY. In 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems. 818–831. Google ScholarDigital Library
- [85] . 2016. An area-efficient FPGA overlay using DSP block based time-multiplexed functional units. arXiv preprint arXiv:1606.06460 (2016).Google Scholar
- [86] . 2015. Savi testbed architecture and federation. In Future Access Enablers of Ubiquitous and Intelligent Infrastructures. Springer, 3–10.Google Scholar
- [87] . 2018. Meltdown: Reading kernel memory from user space. In 27th USENIX Security Symposium (USENIX Security’18). Google ScholarDigital Library
- [88] . 2001. Meeting critical security objectives with security-enhanced Linux. In Ottawa Linux Symposium. 115–134.Google Scholar
- [89] Layong Larry Luo and T. E. G. Tencent. 2018. In Towards Converged SmartNIC Architecture for Bare Metal and Public Clouds at Tencent Scale 2nd Asia-Pacific Workshop on Networking (APNet 2018) August 2-3 2018, Beijing, China.Google Scholar
- [90] . 2020. A hypervisor for shared-memory FPGA platforms. In 25th International Conference on Architectural Support for Programming Languages and Operating Systems. 827–844. Google ScholarDigital Library
- [91] . 2019. An optimal peak hour content server cache update scheduling algorithm for 5G hetnets. In IEEE International Conference on Communications (ICC). 1–6.
DOI: https://doi.org/10.1109/ICC.2019.8761705Google ScholarCross Ref - [92] . 2015. A run time interpretation approach for creating custom accelerators. In 25th International Conference on Field Programmable Logic and Applications (FPL). IEEE, 1–4. Google ScholarDigital Library
- [93] . 2018. In-RDBMS hardware acceleration of advanced analytics. Proc. VLDB Endow. 11, 11 (
July 2018), 1317–1331. Google ScholarDigital Library - [94] . 2014. Virtual eXtensible local area network (VXLAN): A framework for overlaying virtualized layer 2 networks over layer 3 networks. RFC 7348 (2014), 1–22.Google Scholar
- [95] . 2017. LinkBlaze: Efficient global data movement for FPGAs. In International Conference on ReConFigurable Computing and FPGAs (ReConFig). IEEE, 1–8.Google ScholarCross Ref
- [96] . 2021. Domain isolation in FPGA-accelerated cloud and data center applications. In Great Lakes Symposium on VLSI. 283–288. Google ScholarDigital Library
- [97] . 2018. FLexiTASK: A flexible FPGA overlay for efficient multitasking. In Great Lakes Symposium on VLSI. ACM, 483–486. Google ScholarDigital Library
- [98] . 2018. FPGAVirt: A novel virtualization framework for FPGAs in the cloud. In IEEE 11th International Conference on Cloud Computing (CLOUD). IEEE, 862–865.Google ScholarCross Ref
- [99] . 2019. Automatic generation of application-specific FPGA overlays with Rapidwright. In International Conference on Field-Programmable Technology (ICFPT). IEEE, 303–306. Google ScholarDigital Library
- [100] . 2021. Deploying multi-tenant FPGAs within Linux-based cloud infrastructure. ACM Trans. Reconfig. Technol. Syst. 15, 2 (2021), 1–31.Google ScholarDigital Library
- [101] . 2021. Performance study of multi-tenant cloud FPGAs. In IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW). IEEE, 168–171.Google Scholar
- [102] . 2021. A security architecture for domain isolation in multi-tenant cloud FPGAs. In IEEE Computer Society Annual Symposium on VLSI (ISVLSI). IEEE, 290–295.Google Scholar
- [103] . 2020. Architecture support for FPGA multi-tenancy in the cloud. In IEEE 31st International Conference on Application-specific Systems, Architectures and Processors (ASAP). IEEE, 125–132.Google ScholarCross Ref
- [104] . 2019. A hardware-software blueprint for flexible deep learning specialization. IEEE Micro 39, 5 (2019), 8–16.
DOI: https://doi.org/10.1109/MM.2019.2928962Google ScholarCross Ref - [105] . 2008. Implementing an OpenFlow switch on the NetFPGA platform. In 4th ACM/IEEE Symposium on Architectures for Networking and Communications Systems. 1–9. Google ScholarDigital Library
- [106] . 2006. Intel virtualization technology: Hardware support for efficient processor virtualization. Intel Technol. J. 10, 3 (2006).Google ScholarCross Ref
- [107] . 2021. HALF: Holistic auto machine learning for FPGAs. In st IEEE International Conference on Field-Programmable Logic and Applications (FPL).
DOI: https://doi.org/10.1109/FPL53798.2021.00069Google ScholarCross Ref - [108] . 2018. Datacenter traffic control: Understanding techniques and tradeoffs. IEEE Commun. Surv. Tutor. 20, 2 (2018), 1492–1525.
DOI: https://doi.org/10.1109/COMST.2017.2782753Google ScholarCross Ref - [109] . 2016. Distributed denial of service (DDoS) resilience in cloud: Review and conceptual cloud DDoS mitigation framework. J. Netw. Comput. Applic. 67 (
May 2016), 147–165. Google ScholarDigital Library - [110] . 2018. FLOEM: A programming system for NIC-accelerated network applications. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI’18). 663–679. Google ScholarDigital Library
- [111] . 2021. EVEREST: A design environment for extreme-scale big data analytics on heterogeneous platforms. In Design, Automation Test in Europe Conference Exhibition (DATE). 1320–1325.
DOI: https://doi.org/10.23919/DATE51398.2021.9473940Google ScholarCross Ref - [112] . 2018. Bringing FPGAs to HPC production systems and codes. In H2RC’18 Workshop at Supercomputing (SC’18).
DOI: https://doi.org/10.13140/RG.2.2.34327.42407Google Scholar - [113] . 2019. Characterizing power distribution attacks in multi-user FPGA environments. In International Conference on Field Programmable Logic and Applications (FPL). 194–201.Google ScholarCross Ref
- [114] . 2020. Power wasting circuits for cloud FPGA attacks. In International Conference on Field Programmable Logic and Applications (FPL).Google ScholarCross Ref
- [115] . 2014. A reconfigurable fabric for accelerating large-scale data center services. In ACM/IEEE 41st International Symposium on Computer Architecture (ISCA). 13–24.
DOI: https://doi.org/10.1109/ISCA.2014.6853195 Google ScholarDigital Library - [116] . 2021. Pharos: A multi-FPGA performance monitor. In 31st International Conference on Field-Programmable Logic and Applications (FPL). 257–262.
DOI: https://doi.org/10.1109/FPL53798.2021.00048Google ScholarCross Ref - [117] . 2018. FPGA side channel attacks without physical access. In IEEE 26th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM). 45–52.Google Scholar
- [118] . 2019. Leakier wires: Exploiting FPGA long wires for covert-and side-channel attacks. ACM Trans. Reconfig. Technol. Syst. 12, 3 (2019), 11:1–11.29. Google ScholarDigital Library
- [119] . 2015. Security policy enforcement in modern SoC designs. In IEEE/ACM International Conference on Computer-Aided Design. 345–350. Google ScholarDigital Library
- [120] . 2020. Heax: An architecture for computing on encrypted data. In 25th International Conference on Architectural Support for Programming Languages and Operating Systems. 1295–1309. Google ScholarDigital Library
- [121] . 2021. A case for function-as-a-service with Disaggregated FPGAs. In IEEE 14th International Conference on Cloud Computing (CLOUD’21). 333–344.
DOI: https://doi.org/10.1109/CLOUD53861.2021.00047Google ScholarCross Ref - [122] . 2019. System architecture for network-attached FPGAs in the cloud using partial reconfiguration. In 29th International Conference on Field Programmable Logic and Applications (FPL). IEEE, 293–300.
DOI: https://doi.org/10.1109/FPL.2019.00054Google ScholarCross Ref - [123] . 2020. Programming reconfigurable heterogeneous computing clusters using MPI with transpilation. In IEEE/ACM International Workshop on Heterogeneous High-performance Reconfigurable Computing (H2RC). IEEE, 1–9.
DOI: https://doi.org/10.1109/H2RC51942.2020.00006Google Scholar - [124] . 2020. ZRLMPI: A unified programming model for reconfigurable heterogeneous computing clusters. In IEEE 28th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM). IEEE, 220.
DOI: https://doi.org/10.1109/FCCM48280.2020.00051Google Scholar - [125] . 1982. Proof of separability a verification technique for a class of security kernels. In International Symposium on Programming. 352–367. Google ScholarDigital Library
- [126] . 2021. NASCENT: Near-storage acceleration of database sort on SmartSSD. In ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA’21). 262–272. Google ScholarDigital Library
- [127] . 2021. Will ASIC Chips Become the Next Big Thing in AI? Retrieved from https://moorinsightsstrategy.com/will-asic-chips-become-the-next-big-thing-in-ai/.Google Scholar
- [128] . 2018. An inside job: Remote power analysis attacks on FPGAs. In Design, Automation & Test in Europe Conference & Exhibition (DATE). 1111–1116.Google ScholarCross Ref
- [129] . 2012. OpenStack: Toward an open-source solution for cloud computing. Int. J. Comput. Applic. 55, 3 (2012), 38–42.Google ScholarCross Ref
- [130] . 2016. From high-level deep neural models to FPGAs. In 49th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO). IEEE, 1–12. Google ScholarDigital Library
- [131] . 2017. Use of time-series based forecasting technique for balancing load and reducing consumption of energy in a cloud data center. In International Conference on Intelligent Computing and Control (I2C2). 1–6.
DOI: https://doi.org/10.1109/I2C2.2017.8321782Google ScholarCross Ref - [132] . 2005. Virtual Machines: Versatile Platforms for Systems and Processes. Elsevier. Google ScholarDigital Library
- [133] . 2016. FPGA overlays. In FPGAs for Software Programmers. Springer, 285–305. Google ScholarDigital Library
- [134] . 2020. 3D CNN acceleration on FPGA using hardware-aware pruning. In 57th ACM/IEEE Design Automation Conference (DAC). 1–6. Google ScholarDigital Library
- [135] . 2019. Survey of microarchitectural side and covert channels, attacks, and defenses. J. Hardw. Syst. Secur. 3, 3 (
Sept. 2019), 219–234.Google ScholarCross Ref - [136] . 2009. Vulnerability Seen in Amazon’s Cloud-Computing. Retrieved from https://www. technologyreview.com/2009/10/23/208662/vulnerability-seen-in-amazons-cloud-computing/.Google Scholar
- [137] . 2019. libGalapagos: A software environment for prototyping and creating heterogeneous FPGA and CPU applications. In 6th International Workshop on FPGAs for Software Programmers (FSP’19).Google Scholar
- [138] . 2017. Designing for FPGAs in the cloud. IEEE Des. Test 35, 1 (2017), 23–29.Google ScholarCross Ref
- [139] . 2018. Galapagos: A full stack approach to FPGA integration in the cloud. IEEE Micro 38, 06 (
Nov. 2018), 18–24.DOI: https://doi.org/10.1109/MM.2018.2877290Google ScholarCross Ref - [140] . 2017. Enabling flexible network FPGA clusters in a heterogeneous cloud data center. In ACM/SIGDA International Symposium on Field-Programmable Gate Arrays. 237–246. Google ScholarDigital Library
- [141] . 2019. Building the infrastructure for deploying FPGAs in the cloud. In Hardware Accelerators in Data Centers. Springer, 9–33.Google ScholarCross Ref
- [142] . 2021. Retrieved from https://web.archive.org/web/20110904033728http://www.impulseaccelerated.com/.Google Scholar
- [143] . [n.d.]. Quick Start Tutorial for Compiling Deep Learning Models. Retrieved from https://tvm.apache.org/docs/tutorial/relay_quick_start.html.Google Scholar
- [144] . 2021. Cloud FPGA cartography using PCIe contention. In IEEE International Symposium on Field-Programmable Custom Computing Machines.Google Scholar
- [145] . 2019. Temporal thermal covert channels in cloud FPGAs. In International Symposium on Field-Programmable Gate Arrays (FPGA). Google ScholarDigital Library
- [146] . 2020. Fingerprinting cloud FPGA infrastructures. In International Symposium on Field-Programmable Gate Arrays (FPGA). Google ScholarDigital Library
- [147] . 2020. HEAWS: An accelerator for homomorphic encryption on the Amazon AWS FPGA. IEEE Trans. Comput. 69, 8 (2020), 1185–1196.Google Scholar
- [148] . 2020. Trust in FPGA-accelerated cloud computing. Comput. Surv. 53, 6 (
Dec. 2020), 28:1–28:128. Google ScholarDigital Library - [149] . 2019. Efficient PUF-based key generation in FPGAs using per-device configuration. IEEE Trans. VLSI Syst. 27, 2 (
Feb. 2019), 364–375.Google ScholarCross Ref - [150] . 2019. Introducing ReCPRI: A field re-configurable protocol for backhaul communication in a radio access network. In IFIP/IEEE Symposium on Integrated Network and Service Management (IM). 329–336.Google Scholar
- [151] . 2017. PCIeHLS: An OpenCL HLS framework. In 4th International Workshop on FPGAs for Software Programmers. VDE, 1–6.Google Scholar
- [152] . 2019. Multi-FPGA accelerator architecture for stencil computation exploiting spacial and temporal scalability. IEEE Access 7 (2019), 53188–53201.Google ScholarCross Ref
- [153] . 2004. A runtime environment for reconfigurable hardware operating systems. In Field Programmable Logic and Application, , , and (Eds.). Springer Berlin, 831–835.Google Scholar
- [154] . 2020. FPDeep: Scalable acceleration of CNN training on deeply-pipelined FPGA clusters. IEEE Trans. Comput. 69, 8 (2020), 1143–1158.Google ScholarDigital Library
- [155] . 2020. When FPGA meets cloud: A first look at performance. IEEE Trans. Cloud Comput. (2020), 1–1.
DOI: https://doi.org/10.1109/TCC.2020.2992548Google Scholar - [156] . 2006. NetFPGA: A tool for network research and education. In 2nd Workshop on Architectural Research Using FPGA Platforms (WARFP), Vol. 3.Google Scholar
- [157] . 2015. Enabling FPGAs in hyperscale data centers. In IEEE 12th International Conference on Ubiquitous Intelligence and Computing and IEEE 12th International Conference on Autonomic and Trusted Computing and IEEE 15th International Conference on Scalable Computing and Communications and Its Associated Workshops (UIC-ATC-ScalCom). 1078–1086.
DOI: https://doi.org/10.1109/UIC-ATC-ScalCom-CBDCom-IoP.2015.199Google Scholar - [158] . 2016. Disaggregated FPGAs: Network performance comparison against bare-metal servers, virtual machines and Linux containers. In IEEE International Conference on Cloud Computing Technology and Science (CloudCom). 9–17.
DOI: https://doi.org/10.1109/CloudCom.2016.0018Google ScholarCross Ref - [159] . 2016. Network-attached FPGAs for data center applications. In International Conference on Field-Programmable Technology (FPT). 36–43.
DOI: https://doi.org/10.1109/FPT.2016.7929186Google ScholarCross Ref - [160] . 2019. Seiba: An FPGA overlay-based approach to rapid application development. In International Conference on ReConFigurable Computing and FPGAs (ReConFig). IEEE, 1–8.Google ScholarCross Ref
- [161] . [n.d.]. Xilinx Powers Alibaba Cloud FaaS with AI Acceleration Solution for E-Commerce Business. Retrieved from https://www.xilinx.com/publications/powered-by-xilinx/xilinx-alibaba-case-study.pdf.Google Scholar
- [162] Xilinx Corporation 2021. Virtex UltraScale+ FPGA Data Sheet: DC and AC Switching Characteristics.Google Scholar
- [163] . 2018. Hardware acceleration over NFV in China Mobile. OPNFV Plugfest.Google Scholar
- [164] . 2019. A data-center FPGA acceleration platform for convolutional neural networks. In 29th International Conference on Field Programmable Logic and Applications (FPL). 151–158.Google ScholarCross Ref
- [165] . 2020. SoK: On the security challenges and risks of multi-tenant FPGAs in the cloud. arxiv arXiv:2009.13914 (2020).Google Scholar
- [166] . 2021. Trusted configuration in cloud FPGAs. In IEEE International Symposium on Field-Programmable Custom Computing Machines.Google Scholar
- [167] . 2017. The Feniks FPGA operating system for cloud computing. In 8th Asia-Pacific Workshop on Systems. 1–7. Google ScholarDigital Library
- [168] . 2019. Computer organization and design course with FPGA cloud. In 50th ACM Technical Symposium on Computer Science Education. ACM, 927–933. Google ScholarDigital Library
- [169] . 2018. FPGA-based remote power side-channel attacks. In IEEE Symposium on Security and Privacy (S&P). 229–244.Google Scholar
- [170] . 2014. NetFPGA SUME: Toward 100 Gbps as research commodity. IEEE Micro 34, 5 (2014), 32–41.Google ScholarCross Ref
- [171] . 2020. High-density multi-tenant bare-metal cloud with memory expansion SoC and power management. In IEEE Hot Chips 32 Symposium (HCS). 1–18.
DOI: https://doi.org/10.1109/HCS49909.2020.9220447Google Scholar
Index Terms
- The Future of FPGA Acceleration in Datacenters and the Cloud
Recommendations
Deploying Multi-tenant FPGAs within Linux-based Cloud Infrastructure
Cloud deployments now increasingly exploit Field-Programmable Gate Array (FPGA) accelerators as part of virtual instances. While cloud FPGAs are still essentially single-tenant, the growing demand for efficient hardware acceleration paves the way to FPGA ...
Enable Efficient and Flexible FPGA Virtualization for Deep Learning in the Cloud
FPGA '20: Proceedings of the 2020 ACM/SIGDA International Symposium on Field-Programmable Gate ArraysFPGAs have shown great potential in providing low-latency and energy-efficient solutions for deep learning applications, especially for the deep neural network (DNN). Currently, the majority of FPGA based DNN accelerators are designed for single-task ...
Auditing Security Compliance of the Virtualized Infrastructure in the Cloud: Application to OpenStack
CODASPY '16: Proceedings of the Sixth ACM Conference on Data and Application Security and PrivacyCloud service providers typically adopt the multi-tenancy model to optimize resources usage and achieve the promised cost-effectiveness. Sharing resources between different tenants and the underlying complex technology increase the necessity of ...
Comments